Blindness affects the way students learn, with Braille Assist, we hope to make that obstacle easier.
What it does
This machine detects letters from the webcam, using octagon-shaped wheels, they show the words in braille
How we built it
How we built was first we created the program that detected text with our computer's camera. Then we created a program for the arduino that would tell the motors to rotate a specific degree for each character. We then put the arduino and the motors into a box and elevated the motors so octagon could pop out of the hole at the top of the box. Next we combined the two programs and added the featue text-to-speech into our code as well.
Challenges we ran into
One of the main problems we faced was making the dc motors rotate specific degree. Since we did not have stepper motors and dc motors are not designed for this without an encoder, we had to do a lot of math to get us to make it rotate properly. Furthermore, since we used pytesseract it is not known for detecting handwriting, so our project had to be only related to recognizing text with simple fonts like arial.
Accomplishments that we're proud of
The accomplishment that we're proud of is the actual machine itself. It took several hours to complete and hundreds of lines of code.
What we learned
We learned how to use ai with python and how to control arduino using python.
What's next for Braille Assist
We believe that what's next for braille assist is to detect handwriting. With the help of google cloud vision API, we could have achieved this, but because you need a premium tier (it costs money) on google cloud to access it, we couldn't incorporate it into our project. With this, Braille Assist could be used for anything including reading a friend's writing to reading what's on the grocery list.
Log in or sign up for Devpost to join the conversation.